Defect Detection and Depth Estimation in Composite Materials for Pulsed Thermography Images by Nonuniform Heating...

Area of Science:

• Materials Science
• Non-destructive Testing
• Image Analysis

Background:

• Pulsed thermography is a key nondestructive technique for inspecting composite materials.
• Detecting defects in thermal images can be challenging due to low contrast and nonuniform heating.

Purpose of the Study:

• To develop an automated procedure for defect detection in thermal images of composite materials.
• To address challenges posed by low-contrast and nonuniform heating conditions.

Main Methods:

• Utilized pulsed thermography for data acquisition.
• Implemented a novel methodology involving nonuniform heating correction and gradient direction analysis.
• Employed local and global segmentation for analyzing carbon fiber-reinforced plastic (CFRP) thermal images.

Main Results:

• The automated procedure successfully detected defects in CFRP samples with Teflon inserts.
• The method demonstrated reliability under low-contrast and nonuniform heating conditions.
• Achieved superior performance compared to deep learning and background thermal compensation by filtering strategies.

Conclusions:

• The proposed automated defect detection method is effective and robust for pulsed thermography analysis of composite materials.
• The technique offers a simple yet powerful alternative for nondestructive evaluation.
• The method shows significant advantages over existing approaches, particularly in challenging thermal conditions.